• Editorial Board +
• For Contributors +
• Journal Search +
Journal Search Engine
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.19 No.4 pp.866-876
DOI : https://doi.org/10.7232/iems.2020.19.4.866

# The Impact of Financial Risks on the Performance of Russian Banks

Institute of Management, Economics and Finance, Kazan Federal University, Kazan, Russia
*Corresponding Author, E-mail: ellypurwendah@gmail.com
September 20, 2020 October 9, 2020 October 15, 2020

## ABSTRACT

This study uses multiple regression analysis with financial ratios to determine the impact of financial risks and financial leverage on the financial performance of Russian commercial banks in the period 2008-17. This study involves 85 Russian banks, whose total assets comprise 87% of the total assets of the Russian banking sector. The study used six indicators to measure five types of financial risks and an indicator to measure the financial leverage: interest rate risk, foreign exchange risk, liquidity risk, credit risk, operational risk, and financial leverage risk. The study also used three indicators to measure bank performance: Net Interest Margin, Return on Assets, and Return on Equity. The study found that Over the 10 years of the study, the risk contributed to the formation of net interest margin by 87%, return on equity by 50%, and return on assets by 53%. The impact of credit risk, operational risk, and liquidity risks on performance indicators in Russian banks were very positive and significant. The effect of leverage and interest rate risk on performance indicators was negative and very limited. The foreign exchange risks had no effect on performance indicators. The study also found that the other variables (without risk and leverage) had a positive effect on the net interest margin, but their effect on the return on assets, and the return on equity was negative and significant. The study concluded It is important for Russian banks to search for these factors, which led to the negative impact, Russian banks should study these factors and correct their positions.

## 1. INTRODUCTION

Bank failures can lead to systemic crises that have serious consequences for the economy. Thus, good and stable indicators of banks have a positive effect on the economy. But, considering the results of banks and their financial results, it is necessary to take into account the risks faced by banks, because good financial results can cause banks to lose sight of risks that can lead to sudden crises that cannot be fixed. Many studies focus on various aspects of productivity, but they ignore risks, for example, (Luo, 2003). Studying the impact of risk on the bank’s performance is very important because of its effect on long-term profit.

Poor risk management could have serious consequences, so banks need to effectively manage risks. Failure to effectively manage risks in banks can hurt not only bank owners and investors but the whole community. Theoretically, the risk is also the possibility that something will happen, but the word is usually utilized on the negative side, and not on the positive side. Paul and Kourouche (2008) defined risk as an unplanned event leading to financial losses, and Idris and Toh, (2020) defined risk as a situation in which the probability of a result can be quantified and ensured. DeYoung et al. (2001) said that risks are the uncertainty resulting from negative outcomes. Sharma (2003) pointed out that risk is the probability of loss, it may be a financial loss or loss of reputation/image. Kirkwood and Nahm (2006) distinguish between “risk” and “uncertainty”. Uncertainty is the case when the decision-maker knows all the possible results of this action but has no idea about the probability of the results. In contrast, the risk is associated with a situation in which the decision-maker knows the likelihood of different outcomes. In short, the risk is quantitative uncertainty.

Risks are uncertainties that lead to negative fluctuations in profitability or loss. The risk lies at the heart of banking, banking regulation and, ultimately, banking crises. Banks are known as financial intermediaries that borrow money from excess units and lend to deficit units. During this process, they provide four basic services: intermediation in liquidity, intermediation in value, intermediation in risk and intermediation in maturity. This intermediation makes banks face many risks: liquidity risk, operational risk, credit risk, interest rate risk and foreign exchange risk. In general, risks can be divided into two types: systemic risk and non-systemic risk. Systemic risk cannot be controlled by banks because they are created by external factors; non-systemic risk can be controlled to a certain extent by banks because they are produced by internal factors. Since non-systemic risk causes differences in bank performance, bank risk preferences affect bank performance (Sharma, 2003).

Despite the extensive literature on bank performance, banks are often viewed as production or intermediation units, risk possibilities are not closely related to performance. As a rule, in the literature for the analysis of risk factors, only one type of indicator of risk, credit or bad credit is used. Oblad (2019) and Laeven (1999) criticized performance measurement studies for not taking into account the risk-taking activities of banks and suggesting that banks are risk-neutral institutions. Sun and Chang (2011) criticized current research for mainly using credit risk indicators without considering other types of risk to explain effectiveness. Greuning and Bratanovic (2009) noted the importance of risk-based banking analysis that takes into account important factors and relationships in terms of risk assessment.

When comparing the profitability of universal banks with the profitability of banks due to investment activities, the latter were more successful in good times, but they suffered more losses when the crisis erupted. An interesting question here is whether the higher risk was rewarded.

A study by Blair and Heggestad (1978), cited in an article by Hua and Liu (2010), showed the following: “As shown in Figure 1 above, the risk-return frontier appears on the envelope RRF if no risk-free assets are assumed for banks but Not all banks are able to operate their asset portfolios along the frontier. For example, the bank in point A does not take risks efficiently, which could be due to management incapability or operating environment. The bank at point A has many options to improve its efficiency. For instance, it could increase its expected return to point E without changing the risk profile. Meanwhile, the upper bound probability of financial distress also decreases as the z-score ray becomes steeper. If this bank is less conservative, it could move along the original z-score ray (i.e. DZ1) to point D. This scenario would increase its return and its expected risk, but the probability of maximum of financial distress would remain unchanged. If this bank is quite conservative, it may choose to reduce both risk and financial distress probability to point F while maintaining the same expected return. For the bank to be on the frontier, it should trade expected returns for reduced risks (e.g., from point D to E) and has a lower upper bound probability of financial distress.” (Hua and Liu, 2010).

A study by Blair and Heggestad (1978), cited in an article by Hua and Liu (2010), showed the following: “The risk-return profile in Figure 1 may also indicate the financial distress risk of a bank or the probability of a bank under financial distress. Financial distress generally means that a bank cannot pay off, or find it difficulty in paying off, its financial obligations, which can be equivalent to economic insolvency (i.e., the loss exceeds the equity capital). Thus, the financial distress risk can be defined as the probability of the event that a bank’s loss exceeds its total capital. Let $π ˜$ be the random return with mean $E ( π ˜ )$ and the standard deviation $σ π ˜$ Given the portfolio characteristics and the bank’s capital position, Chebyshev’s inequality may indicate the approximate probability of financial distress.” (Riyadi et al., 2019) cited in (Hua and Liu 2010). A study by Blair and Heggestad (1978), cited in an article by Hua and Liu (2010), showed the following:

“Chebyshev’s inequality suggests:

$P { | π ˜ − E ( π ˜ ) | ≥ 2 σ π ˜ } ≤ 1 / z 2$

• Denoting: $E ( π ˜ ) − z σ π ˜ = − k$

one can express the least upper bound probability of bank distress as follows:

$P { π ˜ ≤ − k } ≤ 1 / z 2$

where $1 z = σ π ˜ k + E ( π ˜ ) = σ π ˜ / k 1 + E ( π ˜ k )$ is the reciprocal of the slope of z-score rays (e.g.DZ1, EZ2, and FZ3). This suggests that the flatter the z-score ray, the higher the upper bound.” (Hua and Liu, 2010, p. 4).

A study by Blair and Heggestad (1978), cited in an article by Hua and Liu (2010), showed the following: “This indicates that the flatter the z-score ray, the higher the upper bound probability that a bank would be under financial distress, while along the ray, different risk-return combinations have the same upper bound probability of financial distress (constant z- scores). Banks with strong intrinsic safety and soundness should be able to manage risk efficiently, and therefore, they shall operate close to or even on the risk-return frontier and avoid being exposed to costly financial distress.” (Hua and Liu, 2010: 5).

## 2. METHOD

### 2.1 The Hypotheses

The main hypotheses can be formulated as follows:

• Ho: Risks variables don’t affect financial performance (expressed by NIM, ROA and ROE) of the Russian commercial banks.

• H1: Risks variables affect financial performance (expressed by NIM, ROA, and ROE) in Russian commercial banks.

#### 2.1.1. Subset Hypothesis:

• •NIM Model:

• Ho: (IRR-FXR-LR-CRR-OPR-LEV) don’t affect NIM in Russian banks.

• H1: At least one of the (IRR-FXR-LR-CRR-OPRLEV) affect NIM in Russian banks.

• •ROA Model:

• Ho: (IRR-FXR-LR-CRR-OPR-LEV) don’t affect ROA in Russian banks.

• H1: At least one of the (IRR-FXR-LR-CRR-OPRLEV) affect ROA in Russian banks.

• •ROE Model:

• Ho: (IRR-FXR-LR-CRR-OPR-LEV) don’t affect ROE in Russian banks.

• H1: At least one of the (IRR-FXR-LR-CRR-OPRLEV) affect ROE in Russian banks.

In our quest to find empirical articles on the impact of financial risks on the performance of Russian banks, we managed to find only one article. This article was written by Solanko and Fungáčová (2008). They examined the relationship between the characteristics of Russian banks and risks from 1999 to 2007. Regression analysis was used for grade Z, the results showed that the average risk is still lower than the average in Central and Eastern Europe. Researchers found that large banks have more bankruptcy risks than small banks. In addition, foreign banks had a higher risk of bankruptcy than local banks, and large state-controlled banks are different from other state-controlled banks. They also found that regional banks were more dangerous than their Moscow counterparts. We tried very hard to find articles on the impact of risks on Russian banks, but we did not find articles discussing this topic empirically. Most of the articles that we have found discuss the theoretical aspect of the topic, so we believe that in practice there is not enough research specializing in this topic. We hope that our article will motivate researchers to focus on this topic empirically.

### 2.2 Data & Variables

The purpose of this study is to determine the impact of risk on the performance of Russian banks using financial ratios and multiple regression analysis in the period 2008-2017. This study includes the largest of 85 Russian banks in terms of assets, the total assets of the studied banks formats 87% of the total assets of the banking sector in Russia. Table 1 shows the study variables, its abbreviations, and its calculation methods. Table 2 shows the values of the study variables. All study data were obtained from official data published on the website of the Bank of Russia.

### 2.3 The Multiple-Regression Model

The regression equation that expresses the linear relationships between a single dependent variable and the independent variables is outlined in equation 1:

$Y = α + β 1X1 + β 2X2 + β 3X3 + β 4X4 + β 5X5 + β 6X6 + ε$
(1)

In equation 1, Y is the predicted value of the dependent variable. The values of the independent variables are denoted as X1, X2, X3, X4, X5, X6. α is the constant, β2, β3, β4, β5, β6 are called regression coefficients, finally ε is a random factor. Values are assigned to the constants based on the principle of least squares. NIM, ROA, and ROE are the factors of profitability and performance that are influenced by risk factors (inputs): IRR, FXR, LR, CRR, OPR, and LEV. By putting the study variables in the above equation, three equations can be formed as follows:

$NIM = α + β 1 ( IRR ) + β 2 ( FXR ) + β 3 ( LR ) + β 4 ( CRR ) + β 5 ( OPR ) + β 6 ( LEV ) + ε$
(2)

$ROA = α + β 1 ( IRR ) + β 2 ( FXR ) + β 3 ( LR ) + β 4 ( CRR ) + β 5 ( OPR ) + β 6 ( LEV ) + ε$
(3)

$ROE = α + β 1 ( IRR ) + β 2 ( FXR ) + β 3 ( LR ) + β 4 ( CRR ) + β 5 ( OPR ) + β 6 ( LEV ) + ε$
(4)

## 3. RESULTS

### 3.1 Testing (F) For the Suitability of Study Models

To test the suitability of multiple regression models for analysis, using the distribution test (F-statistics), one of the following hypotheses will be rejected:

• Ho: the model does not fit; when independent variables do not affect dependent variables.

• H1: model fits; when independent variables affect dependent variables.

The decision rule as follows:

Accept Ho If p-value (Sig. F) > 0.05, Accept H1 If p-value (Sig. F) ≤ 0.05

from the results of the analysis in table 4, the following results follow:

• • All models except the both models (20) and (22) : values of p-value (Sig. F) ≤ 0.05, So we shall refuse the null hypothesis H0 and accept the alternative hypothesis H1 , that means At the α = 0.05 level of significance, there is enough evidence to conclude that the predictors are useful for predicting the NIM or ROA or ROE ; therefore, the models are suitable.

• • The models (20) and (22): values of p-value (Sig. F) equal to (0.128) and (0.090) respectively, the both values > 0.05, So we shall accept the null hypothesis H0, that means At the α = 0.05 level of significance, there is not enough evidence to con- clude that at least one predictor is useful for predicting the NIM in model (22) and ROA in model (20); therefore, the models are unsuitable. Therefore, NIM in 2015 and ROA in 2014 will be excluded.

### 3.2 R-Square for The Appropriate Models

The R-square measures the strength of the relationship between the model and the dependent variable. However, this is not a formal relationship check. The Ftest of general significance should test the hypothesis of this relationship. If the f-criterion is significant, we can conclude that the r-square is not equal to zero, and the correlation between the model and the dependent variable is statistically significant.

Table 5 shows the percent variability of independent variables. (square R) shows the relationship between dependent and independent variables, while (R) represents the square root of (R). a value of (R) indicates how independent variables relate to NIM, ROA, and ROE. in addition (adjusted square R) mentions the statistical reduction of variable risks. in other words, (adjusted square R) refers to the compatibility of independent variables with dependent ones, in order to confirm the correctness of decisions based on the regression model.

### 3.3 Testing (T) For the Appropriate Models

To examine the suitability of the multiple regression models for analysis, by using the distribution (T-statistic) test, one of the following hypotheses will be rejected:

Ho: The model is not suitable (when the independent variables don’t affect the dependent variables).

• H1: The model is suitable (when the independent variables do affect the dependent variables).

The decision rule as follows:

Accept Ho If p-value (Sig. T) > 0.05, Accept H1 If p-value (Sig. T) ≤ 0.05.

Table 6 presents the accepted variables in the alter- native hypothesis H1 only. We avoided mentioning the accepted variables in the null hypothesis Ho, because it will be excluded from the DEA analysis. In table 6 all the accepted models in the alternative hypothesis (H1) because all of the p-values ≤ 0.05, so we shall refuse the null hypothesis (Ho) and accept the Alternative Hypothesis (H1). So, At the α = 0.05 level of significance, there exists enough evidence to conclude that the slope (B) of the variables mentioned above is not zero and, hence, that variables are useful as a predictor of NIM, ROA and ROE in Russian banks.

The value of slope B in the table 6 represents the ratio of effect and the type of relationship between the independent variables and the dependent variable. In order to know the importance of risk indicators and its impact on performance indicators, it is necessary to determine its real value compared to all variables. Therefore, we multiply the value B by the mean of the dependent variables, this make us know the value of its effect as compared to other variables.

Table 6 shows the variables accepted only in the alternative hypothesis H1 where are all values of p ≤ 0.05, so we abandon the null hypothesis Ho and accept the alternative hypothesis H1. Thus, at the significance level α = 0.05, there is enough evidence to conclude that the slope (B) of the above variables is not equal to zero and, therefore, these variables are useful as predictors of NIM, ROA, and ROE in Russian banks.

The slope value B in table 6 represents the relationship between the effect and the type of relationship between the independent and dependent variables. To find out the importance of risk indicators and their impact on performance indicators, it is necessary to determine their real value compared to all variables. Therefore, we multiply the value of B by the average value of the dependent variables, which allows us to find out the value of its effectiveness compared to other variables, in Figure 2 this effect is shown in detail.

Figure 2 below shows the effect of risk, leverage, and other variables on performance indicators as a percentage of the formation of performance indicators. Based on the results of the multiple regression analysis, the following can be observed:

• 1. The average annual interest rate margin (NIM) in Russian banks was 5% during the period (2008- 2017), the risk and leverage indicators contributed to forming a net interest margin (NIM) of 87% as follows:

• • Credit risk had a positive effect on the net interest margin (NIM) by 43%, its impact was the largest impact among other indicators, this effect was evident in the period 2011-2014. In 2011 there was a decrease in the interest rate in Russia to 8.6% after the interest rate was 10.9% in 2010. This led to a 32% increase in credit, and the banks' financial results increased by 40%, resulting in a 97% increase in profits after tax. In 2012, the interest rate did not change significantly, as it was raised to 8.15%, but the credit size increased by 20%. In 2013, interest rates decreased to 7.9%, but also the credit size increased by 22%. in 2014 Credit also grew significantly by 33%.

• • The operational risks positively affected the net interest margin (NIM) by 4%.

• •The interest rate risk negatively affected the net interest margin (NIM) by -1%, this effect was from 2012-2013, In 2012 the size of deposits and the size of loans were almost equal. In 2013, the size of deposits became greater than the size of loans in Russian banks

• •Leverage negatively affected the net interest margin (NIM) by -1%.

• •Other variables (other than risk and leverage) had a positive impact on the net interest margin (NIM) by 13%. It can be seen that the effect of other variables was negative in the period 2011-2014, but the positive effect in the other years was greater than the negative impact.

• •The liquidity and foreign exchange risks did not have any effect on the net interest margin (NIM).

• 2. The average annual return on assets (ROA) in Russian banks was 0.3% during the period (2008-2017). The risk indicators contributed to forming the return on assets (ROA) by 53%, as follows:

• •Credit risk had a positive impact on the return on assets (ROA), by 22%.

• •The operational risks positively affected the return on assets (ROA) by 22%.

• •Liquidity risk had a positive impact on the return on assets (ROA), by 9%.

• •Other variables (other than risk and leverage) had a significant negative impact on the return on assets (ROA), at -47%.

• •Leverage, foreign exchange risk, and interest rate risk did not have any impact on the return on assets (ROA).

• 3. The average annual return on equity (ROE) in Russian banks was -0.3% during the period (2008-2017). The risk indicators contributed to the formation of the return on equity (ROE), by 50%, as follows:

• •The operational risks positively affected the return on equity (ROE), at 35%. It can be seen that the impact of operational risks was only negative in 2017, while in the other years was positive. In 2017, deposits were greater than loans in Russian banks, and provisions for loan losses increased by 23%, net profit decreased significantly by -18% in Russian banks, which is the largest decrease after its decrease in 2009. Also, profits after taxes decreased to its largest decrease, reaching -119%.

• •Credit risk had a positive impact on the return on equity (ROE), by 15%. The impact of credit risk on return on equity (ROE) was only in 2017, but it was significant, but in other years, Credit risk had no effect on the return on equity.

• •Other variables (other than risk and leverage) negatively affected the return on equity (ROE), by 50%.

• •Leverage, liquidity risk, foreign exchange risk, and interest rate risk did not have any impact on the return on equity.

After measuring the effect of risk and leverage on performance indicators in Russian banks, it can be said that risk and leverage contributed to the formation of a net interest margin (NIM) by 87%, return on equity (ROE) by 50%, and return on assets (ROA) by 53%. The impact of credit risk and operational risk on performance indicators in Russian banks was very positive and significant. Also, the effect of liquidity risks on the return on assets (ROA) was positive, this reflects the importance of the effect of these risks specifically on the performance of Russian banks, this also indicates that these risks are well managed by Russian banks. As for the effect of leverage and interest rate risk on performance indicators, their impact was negative and very limited. As for the foreign exchange risks, they had no effect.

On the other hand, the other variables (without risk and leverage) had a positive effect on the net interest margin (NIM), but their effect on the return on assets (ROA) and the return on equity (ROE) was negative and significant, by -47% and -50%, respectively, this was a negative impact, It is important for Russian banks to search for these factors, which led to this negative impact, Russian banks should study these factors and correct their positions.

## 4. CONCLUSIONS

This study uses multiple regression analysis with financial ratios to determine the impact of financial risks and financial leverage on the financial performance of Russian commercial banks in the period 2008-17. This study involves 85 Russian banks, whose total assets comprise 87% of the total assets of the Russian banking sector. The study used six indicators to measure five types of financial risks and an indicator to measure the financial leverage: interest rate risk (IRR), foreign exchange risk (FXR), liquidity risk (LR), credit risk (CRR), operational risk (OPR) and financial leverage risk (LEV). The study also used three indicators to measure bank performance: Net Interest Margin (NIM), Return on Assets (ROA), and Return on Equity (ROE).

By using multiple regression analysis, the following results were obtained as a result of the study:

•Over the 10 years of the study, the risk contributed to the formation of a net interest margin (NIM) by 87%, return on equity (ROE) by 50%, and return on assets (ROA) by 53%.

• •The impact of credit risk and operational risk on performance indicators in Russian banks was very positive and significant. Also, the effect of liquidity risks on the return on assets (ROA) was positive, this reflects the importance of the effect of these risks specifically on the performance of Russian banks, this also indicates that these risks are well managed by Russian banks.

• •The effect of leverage and interest rate risk on performance indicators was negative and very limited.

• •The foreign exchange risks had no effect on performance indicators

• •The other variables (without risk and leverage) had a positive effect on the net interest margin (NIM), but their effect on the return on assets (ROA) and the return on equity (ROE) was negative and significant, by -47% and -50%, respectively, this was a negative impact, It is important for Russian banks to search for these factors which led to this negative impact, Russian banks should study these factors and correct their positions.

## Figure

Risk-return profile.

The multiple -regression results. the average effect of risk, leverage and other variables on performance indicators in Russian banks (2008-2017).

## Table

Variables definition & measurement units

Values of the study variables

List of banks studied, 2017 (Millions RUB)

ANOVA - F- Test (2008-17)

Total divergence in the dependent variables, (2008-17)

T-Test (2008-17)

## REFERENCES

1. Blair, R. D. and Heggestad, A. A. (1978), Bank portfolio regulation and the probability of bank failure: Note, Journal of Money, Credit, and Banking, 10(1), 88-93.
2. DeYoung, R. E. , Hughes, J. P. , and Moon, C. G. (2001), Efficient risk-taking and regulatory covenant enforcement in a deregulated banking industry, Journal of Economics and Business, 53(2-3), 255-282.
3. Greuning, H. V. and Bratanovic, S. B. (2009), Analyzing Banking Risk: A Framework for Assessing Corporate Governance and Risk Management (3rd Ed.), The World Bank, Washington, DC, 442
4. Hua, C. and Liu, L. (2010), Risk-return efficiency, financial distress risk, and bank financial stress ratings, Asian Development Bank Institute(Adbi),Working Paper 240. Tokyo, Available from: http://www.adbi. org/workingpaper/2010/08/11/4036.
5. Idris, I. and Toh, E. B. H. (2020), Mass media effects in formation of attitudes towards an internalized shopping center: The Conceptual understanding on the usefulness of advertising media, International Journal of Criminology and Sociology,
6. Kirkwood, J. and Nahm, D. (2006), Australian banking efficiency and its relation to stock returns, Economic Record, 82(258), 253-267.
7. Laeven, L. A. (1999), Risk and efficiency in east Asian Banks, World Bank Policy Research, Working Paper No. 2255.
8. Luo, X. (2003), Evaluating the profitability and marketability efficiency of large banks: An application of data envelopment analysis, Journal of Business Research, 56(8), 627-635.
9. Oblad, T. (2019), Cyberbullying among emerging adults: Exploring prevalence, impact, and coping methods, International Journal of Criminology and Sociology, 8, 45-54.
10. Paul, S. and Kourouche, K. (2008), Regulatory policy and the efficiency of the banking sector in Australia, Australian Economic Review, 41(3), 260-271.
11. Riyadi, B. S. , Wibowo, B. R. , and Susanti, V. (2019), Culture of corruption politicians’ behavior in parliament and state official during reform government Indonesia (Genealogical study), International Journal of Criminology and Sociology, 9, 52-62.
12. Sharma, H. S. (2003), Risk management in banks-emerging issues, Banking Finance, 2(6), 5-6.
13. Solanko, L. and Fungáčová, Z. (2008), Risk taking by Russian Banks: Do location, ownership & size matter? Bofit, Discussion Paper No. 21, 41.
14. Sun, L. and Chang, T. (2011), A comprehensive analysis of the effects of risk measures on bank efficiency: Evidences from emerging Asian countries, Journal of Banking and Finance, 35(7), 1727-1735.